PhaseBridge — Phase Interchange Format (PIF)
A strict, lossless bridge between discrete alphabets and continuous phase space.
Status: reference implementation
Overview
PhaseBridge (PIF) defines a minimal, verifiable data representation that maps discrete alphabets (uintM) to a continuous phase domain θ ∈ [0, 2π) and back — losslessly under the declared alphabet. PIF packages a small JSON schema, optional MessagePack serialization, integrity checks (SHA-256), and a single diagnostic coherence metric κ that is meaningful across time series and images.
Why PhaseBridge?
- Multi-modal pipelines mix formats (tables, images, signals) and need a lightweight, auditable interchange layer.
- Existing containers/codecs focus on storage/compression; learned embeddings sacrifice invertibility.
- PhaseBridge provides strict round-trip guarantees, simple integrity verification, and a unified diagnostic signal — ideal where reversibility and trust matter.
Key benefits
- Bit-exact round-trip when
meta.note = "no_processing". - Integrity via optional
meta.hash_raw(SHA-256). - Universal diagnostic κ (coherence) usable across modalities.
- Lightweight integration: reference Python SDK, CLI tools, JSON Schema.
- Conformance tests and property-based round-trip checks included.
How it works — the minimal flow
Declare alphabet — define schema.alphabet.M.
Encode — map discrete symbol n ∈ {0..M−1} to θ = 2π·n/M (stored as float64).
Package — produce a PIF record (schema, θ, optional amplitude, meta, optional hash).
Validate —
pb-validatechecks format, hash and κ diagnostics.Decode —
pb-decodereconstructs discrete symbols exactly (if no-processing mode used).
Project status
Repository: https://github.com/synqratech/phasebridge — reference implementation, CLI, tests.
Publication: preprint archived on Zenodo: https://zenodo.org/records/17257115.
Current state: Working reference code, conformance & property tests, example scripts. Ready for exploration and integration; means-of-contribution available via issues/PRs.
When to use PhaseBridge (use cases)
You need strict, reversible representation for archival or verification.
You require run-time integrity checks at integration boundaries.
You want a single, modality-agnostic diagnostic (κ) for monitoring structural coherence.
Research prototypes, cognitive pipelines, and experimental multimodal stacks where invertibility matters.
Limitations & notes
Round-trip guarantees assume IEEE-754 float64 and declared alphabet constraints (M ≤ 2³²).
PhaseBridge is not a compression codec for storage savings — it is a representational substrate focused on reversibility and diagnostics.
Do not use for already lossy or corrupted pipelines unless data are explicitly marked as processed (
meta.note).
Repo contents & architecture (what you’ll find)
S1PhaseCodec — core n↔θ codec for lossless mapping.
PIF dataclass & JSON Schema — canonical record format (v1).
CLI tools:
pb-encode,pb-decode,pb-validate,pb-kappa.Conformance suite — property-based tests and example round-trip cases.
Examples — image/time series round trips and kappa monitoring.
Contributing & community
GitHub: https://github.com/synqratech/phasebridge — issues, PRs, test contributions welcome.
Preprint: https://zenodo.org/records/17257115 — formal description and experiments.
Contact / coordination: open issues on GitHub or email info@synqra.tech for integration requests.
License & releases
Check repository root for license (MIT). We recommend tagging releases and archiving on Zenodo for DOI/long-term citation (Zenodo already hosts the preprint).
Bring explainable intelligence with NSI
We are opening pilot projects in finance, telecom, industry, healthcare — and beyond. NSI delivers real-time semantic alerts, explainable pattern libraries, and adaptive system maps.